Untitled

ONLINE FIRST
Postoperative Antibacterial Prophylaxis
for the Prevention of Infectious Complications
Associated With Tube Thoracostomy in Patients
Undergoing Elective General Thoracic Surgery
A Double-blind, Placebo-Controlled, Randomized Trial
David A. Oxman, MD; Nicolas C. Issa, MD; Francisco M. Marty, MD; Alka Patel, PharmD; Christia Z. Panizales, BS;Nathaniel N. Johnson, BS; J. Humberto Licona, MD; Shannon S. McKenna, MD; Gyorgy Frendl, MD, PhD;Steven J. Mentzer, MD; Michael T. Jaklitsch, MD; Raphael Bueno, MD; Yolonda Colson, MD, PhD;Scott J. Swanson, MD; David J. Sugarbaker, MD; Lindsey R. Baden, MD Objective: To determine whether extended postopera-
Results: A total of 245 patients were included in the modi-
tive antibacterial prophylaxis for patients undergoing elec- fied intention-to-treat analysis (121 in the intervention tive thoracic surgery with tube thoracostomy reduces the group and 124 in the placebo group). Thirteen patients risk of infectious complications compared with preop- (10.7%) in the intervention group and 8 patients (6.5%) in the placebo group had a primary end point (risk dif-ference, Ϫ4.3% [95% CI, Ϫ11.3% to 2.7%]; P=.26). Six Design: Prospective, randomized, double-blind, placebo-
patients (5.0%) in the intervention group and 5 patients (4.0%) in the placebo group developed surgical site in-fections (risk difference, Ϫ0.93% [95% CI, Ϫ6.1% to Setting: Brigham and Women’s Hospital, an 800-bed ter-
4.3%]; P = .77). Seven patients (5.8%) in the interven- tiary care teaching hospital in Boston, Massachusetts.
tion group and 3 patients (2.4%) in the placebo groupdeveloped pneumonia (risk difference, Ϫ3.4% [95% CI, Participants: A total of 251 adult patients undergoing
Ϫ8.3% to 1.6%]; P=.21). One patient in the interven- elective thoracic surgery requiring tube thoracostomy be-tween April 2008 and April 2011.
tion group developed empyema. No patients experi-enced C difficile colitis.
Interventions: Patients received preoperative antibac-
terial prophylaxis with cefazolin sodium (or other drug
Conclusions: Extended postoperative antibacterial pro-
if the patient was allergic to cefazolin). Postoperatively, phylaxis for patients undergoing elective thoracic sur- patients were randomly assigned (at a 1:1 ratio) using a gery requiring tube thoracostomy did not reduce the num- computer-generated randomization sequence to receive ber of infectious complications compared with extended antibacterial prophylaxis (n = 125) or placebo (n=126) for 48 hours or until all thoracostomy tubes wereremoved, whichever came first.
Trial Registration: clinicaltrials.gov Identifier:
NCT00818766
Main Outcome Measures: The combined occur-
Author Aff
rence of surgical site infection, empyema, pneumonia, JAMA Surg. Published online January 16, 2013. and Clostridium difficile colitis by postoperative day 28.
erative prophylactic antibacterials are rou- tinely given for 24 or 48 hours after the placement of the chest tube and often for See Invited Critique
rently, the approach to postoperative an- Author Affiliations are listed at
tibacterial prophylaxis for these patients varies widely among clinicians.1,2 Postop- goal,6 but the unnecessary use of antibac- 2013 American Medical Association. All rights reserved.
Downloaded From: http://archsurg.jamanetwork.com/ by a Harvard University User on 01/16/2013
terials is costly, may be associated with increased anti- STUDY PROCEDURES AND OUTCOMES
microbial resistance and toxicity, and could be compli-cated by drug-resistant bacterial infections including All of the patients who enrolled in our study received the stan- Clostridium difficile colitis, which could be deadly.7,8 Pre- dard skin preparation (povidone-iodine) and the standard rec- operative surgical prophylaxis using first-generation ommended doses of preoperative antibacterial therapy to pre-vent surgical site infection according to institutional guidelines.
cephalosporins administered within 30 minutes before The patients with a prolonged operative time received addi- the incision time is currently recommended to prevent tional antibacterial doses intraoperatively. Immediately after sur- surgical site infections in clean or clean-contaminated op- gery, patients were randomly assigned to either additional post- erations.6 Additional intraoperative antibacterials could operative antibacterials (intervention group) or placebo.
also be administered for lengthy surgical procedures.9 To Postoperatively, patients received either additional antibacte- assess whether additional antibacterials administered post- rials or placebo for 48 hours or until all chest tubes were re- operatively to patients undergoing elective thoracic sur- moved, whichever occurred first. Patients with no history ofan allergic reaction to cephalosporins received cefazolin (1 g gery with thoracostomy tube placement leads to a re- for patients weighing Յ80 kg and 2 g for patients weighing Ͼ80 duced rate of infectious complications, we compared 2 kg) or placebo every 8 hours. Patients with a history of an al- approaches for the prevention of surgical site infec- lergic reaction to ␤-lactams received 1 g of vancomycin or pla- tions. The first approach consists of administering stan- cebo every 12 hours. Chest tube management was done by the dard preoperative antibacterial prophylaxis only, with no surgical team as per routine postprocedure care.
additional antibacterial therapy given postoperatively. The The primary study end point was the combined incidence second approach also uses standard preoperative anti- of the following postoperative infectious complications: sur- bacterial prophylaxis, but with additional antibacterial gical site infection, empyema, pneumonia, and the occurrenceof C difficile colitis within 28 days of surgery. Definitions of the therapy given for 48 hours postoperatively or until all infectious end points are based on criteria from the Centers for thoracostomy tubes are removed, whichever comes first.
Disease Control and Prevention and are listed in Table 1. The
protocol’s prespecified secondary end points were the length
of hospital stay, the time to removal of the chest tubes, the need
for a reoperation, the administration of additional antibacteri-als for any reason within 28 days after surgery, any allergic re-actions, and 28-day all-cause mortality. Surveillance for study STUDY DESIGN AND PARTICIPANTS
end points was performed daily while the patient was hospi-talized, at the patient’s 2-week postoperative clinic appoint- We conducted this randomized, double-blind, placebo- ment (at which time all patients received routine chest radi- controlled study at Brigham and Women’s Hospital, an 800- ography and evaluation by their surgeon), and by a telephone bed tertiary care teaching hospital in Boston, Massachusetts.
Our study was approved by the hospital’s institutional review Each potential infectious complication was adjudicated by board. Patients 18 years or older presenting for elective lung 2 study investigators. If there was disagreement between the 2 surgery who were expected to require tube thoracostomy were study investigators on the outcome classification, a third in- approached for participation in our study at the time of pre- vestigator decided the final outcome after reviewing the data operative evaluation. Patients gave informed written consent and following the preset end-point definitions. All end points prior to enrollment. Patients were ineligible to participate if any and postrandomization data were verified and adjudicated be- of the following procedures were planned: pneumonectomy, decortication, pleurodesis or pleurectomy, lung volume reduc-tion, esophagectomy, or lung transplantation. Patients with STATISTICAL ANALYSIS
known active infections prior to surgery, with cystic fibrosis,or with a history of antimicrobial use within 1 week of the sur- The reported occurrences of surgical site infection, empyema, gery were excluded. Patients with an estimated glomerular fil- and postprocedure pneumonia following elective thoracic sur- tration rate of less than 60 mL/min or with an allergic reaction gery range from 7% to 14%.3,4,12-25 We estimated that we would to both ␤-lactam antibiotics and vancomycin hydrochloride were need to enroll 352 patients with a power of 80% and a 1-sided ␣ level of .05 to exclude more than 3% absolute difference be-tween the treatment groups. The projected accrual period was RANDOMIZATION AND MASKING
3 years, and no interim analysis was planned. The study ac-crual was stopped at 3 years owing in part to planned changesin preoperative methicillin-resistant Staphylococcus aureus Eligible patients were randomly assigned (at a 1:1 ratio) to re- screening and owing to funding constraints. Data were ana- ceive either normal saline (placebo arm) or additional antibac- lyzed using SAS 9.2 software (SAS Institute). The Fisher exact terial treatment (cefazolin or vancomycin for cephalosporin- test and the Wilcoxon rank sum test were used to compare bi- allergic patients) immediately following surgery. These patients nary variables and continuous variables, respectively. Logistic were randomly assigned in blocks of 4 via a randomization table regression analysis was used to adjust for potential confound- that was prepared for our study, and patients were assigned to study arms sequentially by the members of the hospital’s In-vestigational Drug Service, who were unaware of patient char-acteristics. Study drug doses were prepared by the Investiga- tional Drug Service and administered in intravenous bagsidentical in appearance to those used in clinical practice. Nurses,clinicians, study investigators, and patients were masked to as- Between April 4, 2008, and April 4, 2011, a total of 292 signed study group throughout the study and were unaware patients consented to participate in our study and were of the block randomization strategy.
assessed for eligibility. Forty-one patients were ex- 2013 American Medical Association. All rights reserved.
Downloaded From: http://archsurg.jamanetwork.com/ by a Harvard University User on 01/16/2013
Table 1. Definitions of Infectious End Points
Infectious End Point
Surgical site infections
Superficial surgical site infection involves only skin or subcutaneous tissue around incision and has at least 1 of the following criteria: Organisms isolated from aseptically obtained culture Pain or tenderness, localized swelling, redness or heat, and the incision is deliberately opened by surgeon.
Diagnosis of a superficial wound infection by surgeon Deep surgical site infection involves deep soft tissues, such as fascia or muscle, and has at least 1 of the following: Purulent drainage from the incision but not from the Deep incision spontaneously dehisces or is deliberately opened by surgeon when the patient has at least 1 of the followingsigns or symptoms: fever (Ͼ38°C), localized pain, ortenderness.
An abscess or other evidence of infection involving the incision is found on direct examination (ie, histopathologic or Diagnosis of a deep wound infection by surgeon Positive pleural culture result or purulence within the thoracic space A new infiltrate on chest radiograph associated with at least 3 of the Fever (Ͼ38°C)Purulent sputumLeukopenia (WBC count of Ͻ4000/µL) or leukocytosis (WBC Sputum culture with pathogenic bacteriaIncreased oxygen requirements Positive for C difficile toxin assay results and any 1 of the following: Leukopenia (WBC count of Ͻ4000/µL) or leukocytosis (WBC Findings from sigmoidoscopy, colonoscopy, or histopathologic examination consistent with C difficile infection Abbreviation: WBC, white blood cell.
SI conversion factor: To convert WBC count to ×109 per liter, multiply by a Definitions obtained from Garner et al10 and Gerding et al.11 cluded (Figure). Of those who were excluded, the most
common reasons for exclusion were cancellation of sur-
Figure. Patient flowchart showing the enrollment and randomization of 292
patients for a prospective, randomized, double-blind, placebo-controlled trial
gery (14 patients [34.1%]), receipt of antibacterials other to determine whether extended postoperative antibacterial prophylaxis for than preoperative prophylactic antibiotics (7 patients patients undergoing elective thoracic surgery with tube thoracostomy [17.1%]), and a procedure not meeting inclusion crite- reduces the risk of infectious complications compared with preoperativeprophylaxis only. GFR indicates glomerular filtration rate.
A total of 251 patients were randomly assigned post- operatively to receive either extended antimicrobial pro- ment in our study were analyzed. All patients com- phylaxis (intervention group; n = 125) or no additional pleted the 2-week visit. We were unable to contact 21 antibacterial prophylaxis (placebo group; n = 126). Of the patients (11 in the intervention group and 10 in the pla- 125 patients allocated to receive extended prophylaxis, cebo group) for the day 28 follow-up telephone call. One 3 did not receive the allocated intervention: one with- patient in the intervention group died of complications drew consent, one was found to have an infection dur- of atrial fibrillation and cerebrovascular accident before ing surgery, and one erroneously received a nonstudy open-label antibacterial after surgery. All patients allo- A total of 245 patients were included in the final analy- cated to receive placebo received the allocated interven- sis (121 in the intervention group and 124 in the pla- tion. Three patients (1 in the intervention group and 2 cebo group). Sixteen patients had protocol violations af- in the placebo group) were enrolled twice in our study ter being randomly assigned (7 in the intervention group at separate occasions. Only data from their first enroll- and 9 in the placebo group): 6 patients received addi- 2013 American Medical Association. All rights reserved.
Downloaded From: http://archsurg.jamanetwork.com/ by a Harvard University User on 01/16/2013
Table 2. Baseline Characteristics of Intention-to-Treat Population
Patients, No. (%)
Intervention Arm
Placebo Arm
Characteristic
P Value
Time to antibacterial administration prior to incision,d median (range) min Additional intraoperative doses of antibacterial administered Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); VATS, video-assisted thoracoscopic surgery.
a One patient received levofloxacin and metronidazole hydrochloride in addition to cefazolin.
b Six patients received clindamycin, and 1 patient received levofloxacin.
c Three patients received clindamycin, and 1 patient received levofloxacin and metronidazole.
d Incision time was not recorded for 9 patients (ie, 114 in the intervention arm and 122 in the placebo arm).
tional doses of the study drug after removal of the chest ministered was similar in both arms (median, 5 doses tube (3 in each arm), 4 patients missed 1 study drug dose [range, 0-6 doses]; P = .70).
(3 in placebo group and 1 in intervention group), 3 pa- Thirteen patients had at least 1 or more events as part tients had a study drug dose delayed by 2 hours or more of the primary study end point in the intervention group (2 in the placebo group and 1 in the intervention group), compared with 8 patients in the placebo group (risk dif- 2 patients inadvertently received 1 dose of wrongly as- ference [RD], Ϫ4.3% [95% CI, Ϫ11.3% to 2.7%]; P = .26) signed study drug (1 in each arm), and 1 patient in the (Table 4). Six patients had surgical site infections in the
intervention group received 1 lower dose of cefazolin.
intervention group compared with 5 patients in the pla- The patients’ baseline characteristics are summarized cebo group (RD, Ϫ0.9% [95% CI, Ϫ6.1% to 4.3%]; in Table 2. The median age and body mass index and
P = .77). Seven patients were diagnosed with pneumo- the proportions of patients with regard to sex and race were nia in the intervention group compared with 3 patients similar in both arms. The median duration of operation, in the placebo group (RD, Ϫ3.4% [95% CI, Ϫ8.3% to the type of surgery, the type of antibacterial prophylaxis 1.6%]; P = .21). One patient in the intervention group used, the median time to antibacterial administration prior had an empyema. This patient also had a surgical site in- to incision, and the number of additional intraoperative fection. No patient in either group developed C difficile doses of antibacterials administered were all similar in both colitis or an allergic reaction to a drug.
groups. The type of surgical approach (video-assisted tho- Of the 245 patients included in the final analysis, 63 racoscopic surgery [VATS] vs open thoracotomy) was simi- (26.0%) received additional antibacterials postopera- lar in both groups. Wedge resection was the most com- tively during the study period (32 in the intervention monly performed surgical intervention (146 of 245 patients group compared with 31 in the placebo group). Of these [60.0%]), followed by lobectomy (76 of 245 patients 63 patients, 21 (33.3%) received additional antimicro- [31.0%]). Cefazolin was used for preoperative prophy- bials because they had one of the study outcomes (sur- laxis in 227 of the 245 patients (93.0%).
gical site infection, pneumonia, or empyema). The most Postoperatively, the most common pathologic find- common indications for additional antimicrobials dur- ing was malignancy (76.0%) (86 in the intervention group ing the study period, excluding patients with study end vs 100 in the placebo group). The number of postopera- points, were urinary tract infection (15 patients [6.1%]), tive chest tubes placed and the median time to removal upper respiratory tract infection (13 patients [5.3%]), and of all chest tubes were similar in both arms (Table 3).
additional surgical interventions (9 patients [3.7%]) The number of postoperative doses of study drug ad- (Table 3). The median time to use of additional antibac- 2013 American Medical Association. All rights reserved.
Downloaded From: http://archsurg.jamanetwork.com/ by a Harvard University User on 01/16/2013
Table 3. Postrandomization Findings in Intention-to-Treat Population
Patients, No. (%)
Intervention Arm
Placebo Arm
P Value
Additional antibacterials Ͻ28 d after randomization Indication for additional antibacterials during follow-up Time to additional antibacterials after surgery for patients without study end Abbreviations: IQR, interquartile range; PCP, Pneumocystis jiroveci pneumonia.
a One patient had tuberculosis, 1 had aspergillosis, and 1 had histoplasmosis.
Table 4. Primary Study End Point Results
Patients, No. (%)
Intervention Arm
Placebo Arm
End Point
Risk Difference (95% CI)
P Value
a One patient in the intervention arm had both a deep surgical site infection and an empyema.
terials after surgery for patients without a study end point and skin lesion excision (n = 1). The median length of were 12 days (range, 1-28 days) in the intervention group stay (3 days) was no different between the 2 arms. Nine and 5 days (range, 1-30 days) in the placebo group patients (7.3%) in the placebo arm had upper respira- (P = .31). Fifteen patients (6.1%) required a reopera- tory tract infection compared with 4 patients (3.3%) in tion: 5 in the intervention group and 10 in the placebo the intervention arm. Although these patients did not meet group (P = .29). Among the 15 patients who required a definition criteria for pneumonia, we explored the ef- reoperation, only 1 procedure (decortication) was per- fect of additional postoperative antimicrobial treatment formed for a patient in the intervention group who had on an expanded composite end point that included any empyema. The other 14 reoperations were performed for respiratory tract infection for which patients received ad- the following reasons: vocal cord paralysis (n = 3), per- ditional antibacterial treatment. When any respiratory tract cutaneous gastric tube placement (n = 2), air leak (n = 2), infection for which additional antibacterials were pre- tracheostomy (n = 1), hemothorax (n = 1), chest wall he- scribed were included in the composite outcome, there matoma (n = 1), chest wall necrosis without infection were no differences in the number of outcomes in either (n = 1), adhesiolysis (n = 1), nasal septal deviation (n = 1), group (17 events in each arm; RD, Ϫ0.34% [95% CI, 2013 American Medical Association. All rights reserved.
Downloaded From: http://archsurg.jamanetwork.com/ by a Harvard University User on 01/16/2013
Ϫ9.0% to 8.3%]; P = .99). The number of events was simi- Several studies32,33 have documented that an impor- lar in both groups (6 in the intervention group and 5 in tant proportion of postoperative infectious complica- the placebo group) when pneumonia was excluded from tions develop after the patient has been discharged from the composite end point (RD, Ϫ0.93% [95% CI, Ϫ6.12% the hospital. This makes it challenging to correctly clas- sify and capture study end points after hospital dis-charge. In our study, the compliance rate of patients at2 weeks after surgery was 100%. Patients were physi- cally seen and examined by the operating surgeon, un-derwent chest radiography, and were assessed for the oc- To establish evidence-based standards for surgical pro- currence of any of the study end point. In addition, we phylaxis for patients undergoing elective general tho- were able to contact the majority of patients at postop- racic surgical procedures requiring the placement of a chest erative day 28. We were unable to reach 22 patients (9%) tube, we examined whether additional postsurgical anti- by phone at day 28; however, we were able to assess for bacterial prophylaxis for 48 hours or until all chest tubes occurrence of any study end point by reviewing the docu- were removed, whichever occurred first, decreased fur- mented follow-up visits for all patients.
ther the rate of infectious complications compared with Although our study was closed to enrollment before standard preoperative prophylaxis only. We also exam- reaching its target accrual, the findings demonstrate the ined the possible deleterious effects that could poten- lack of benefit of extended antibacterial prophylaxis for tially be associated with the use of extended antibacterial patients who undergo elective thoracic surgical proce- prophylaxis (ie, allergic reaction and C difficile colitis).
dures that require chest tube placement. More end points The major finding in our study is that extended post- occurred in the intervention arm (10.7%) than in the pla- surgical prophylaxis did not offer any additional ben- cebo arm (6.5%) (RD, Ϫ4.3% [95% CI, Ϫ11.3% to 2.7%]).
efits in terms of reducing the overall rate of postopera- The 95% CI of the RD on the composite end point be- tive infectious complications when compared with tween treatment arms excluded our protocol’s prespeci- placebo. In fact, more composite end points occurred in fied 3% RD in favor of extended antimicrobial prophy- the intervention arm compared with the placebo arm laxis. It is unlikely that increasing the number of patients enrolled would have resulted in a clinically meaningful dif- We explored several important variables that could ference that would make the findings in our study less ap- have skewed the results of our study. All variables were plicable. In addition, the numbers of patients who re- evenly distributed in both arms of the study. The base- ceived additional antimicrobials during the follow-up line characteristics in the intervention arm and the pla- period were similar in both arms. When any respiratory cebo arm were all well balanced, including age, weight, tract infection for which additional antibacterials were pre- body mass index, procedure type, and duration of sur- scribed were included in the composite outcome, there were gery. The majority of patients received their preopera- no differences in the number of outcomes in either group tive prophylactic antimicrobials within 60 minutes of the (17 events in each arm; RD, Ϫ0.34% [95% CI, Ϫ9.0% to incision time (Table 2), as recommended by the Cen- ters for Disease Control and Prevention and the Society The findings in our study might not be applicable to all of Thoracic Surgeons practice guidelines.6,26 types of thoracic surgical procedures. Patients undergo- The estimated occurrence of surgical site infection, em- ing more complex procedures such as pneumonectomy, pyema, and postoperative pneumonia in patients under- lung volume reduction, and esophagectomy were not eli- going elective thoracic surgery varies widely among stud- gible for enrollment in our study. Patients requiring these ies depending on the criteria employed to define infectious types of procedures often have coexisting morbidities that end points, the type of surgery performed (VATS vs open inherently carry an increased risk of infectious complica- thoracotomy), and the preoperative antimicrobial pro- tions, including fistulization, dead space infection, and the phylactic regimen used. In most studies,3,4,12-17,19-25 the es- prolonged placement of drains. Our findings, however, may timated occurrences of surgical site infection, pneumo- be applicable to the vast majority of elective clean or clean- nia, and empyema after elective thoracic surgical contaminated thoracic surgical procedures requiring tube thoracostomy placement. These findings could also be ex- In our study, the overall rate of infectious complica- trapolated to other types of surgery in which a tube or a tions and the rates of surgical site infection, pneumo- drain is inserted in sterile fashion in a sterile compart- nia, and empyema were all within those observed in simi- lar studies involving elective thoracic surgery (8.6%). Of In summary, extending postoperative surgical pro- the 245 patients, 132 (54.0%) underwent a VATS; this phylaxis for patients undergoing elective general tho- procedure has a relatively lower rate of perioperative com- racic surgery with chest tube placement does not seem plications compared with an open thoracotomy.16,18,19,25,27-31 to offer any additional benefits in terms of reducing in- In our study, the number of patients who underwent an fectious complications compared with standard preop- open thoracotomy and the number of patients who un- erative surgical prophylaxis. Although not demon- derwent a VATS were similar in both arms of our study strated in our study, extending nonbeneficial antibacterial (P = .89) (Table 2). More infectious outcomes were noted prophylaxis to the postsurgery period could potentially in the open thoracotomy group (11.5%) compared with be associated with adverse effects such as the selection the VATS group (6.1%), which is consistent with previ- of more drug-resistant organisms, allergic reactions, drug ously published studies.16,18,19,25,27,28,30,31 toxicities, C difficile colitis, and higher costs.
2013 American Medical Association. All rights reserved.
Downloaded From: http://archsurg.jamanetwork.com/ by a Harvard University User on 01/16/2013
Accepted for Publication: October 5, 2012.
5. Otani S, Endo S, Sato Y, Hasegawa T, Saito N, Sohara Y. Feasibility of short- Published Online: January 16, 2013. doi:10.1001
term antibiotic prophylaxis after pulmonary resection [in Japanese]. Kyobu Geka.
2004;57(13):1171-1174, discussion 1175-1176.
6. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR; Centers for Disease Author Affiliations: Divisions of Infectious Diseases (Drs
Control and Prevention (CDC) Hospital Infection Control Practices Advisory Com- Oxman, Issa, Marty, Licona, and Baden) and Thoracic mittee. Guideline for Prevention of Surgical Site Infection, 1999. Am J Infect Control.
Surgery (Drs Mentzer, Jaklitsch, Bueno, Colson, Swan- 1999;27(2):97-132, quiz 133-134, discussion 96.
son, and Sugarbaker, Ms Panizales, and Mr Johnson), De- 7. Greenstein AJ, Byrn JC, Zhang LP, Swedish KA, Jahn AE, Divino CM. Risk fac- tors for the development of fulminant Clostridium difficile colitis. Surgery. 2008; partments of Pharmacy (Dr Patel) and Anesthesiology (Drs McKenna and Frendl), Brigham and Women’s Hospital, 8. Zerey M, Paton BL, Lincourt AE, Gersin KS, Kercher KW, Heniford BT. The bur- Boston, Massachusetts; and Division of Pulmonary and den of Clostridium difficile in surgical patients in the United States. Surg Infect Critical Care, Thomas Jefferson University Hospital, Phila- (Larchmt). 2007;8(6):557-566.
9. Hranjec T, Swenson BR, Sawyer RG. Surgical site infection prevention: how we do it. Surg Infect (Larchmt). 2010;11(3):289-294.
Correspondence: Nicolas C. Issa, MD, Division of Infec-
10. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for noso- tious Diseases, Brigham and Women’s Hospital, 75 Fran- comial infections, 1988. Am J Infect Control. 1988;16(3):128-140.
cis St, PBB A-4, Boston, MA 02115 (nissa@partners.org).
11. Gerding DN, Johnson S, Peterson LR, Mulligan ME, Silva J Jr. Clostridium difficile- Author Contributions: Drs Oxman and Issa contributed
associated diarrhea and colitis. Infect Control Hosp Epidemiol. 1995;16(8):459-477.
equally to this work. Dr Issa had full access to all the data 12. Boldt J, Piper S, Uphus D, Fu¨ssle R, Hempelmann G. Preoperative microbio- in the study and takes responsibility for the integrity of logic screening and antibiotic prophylaxis in pulmonary resection operations. Ann the data and the accuracy of the data analysis. Study con- Thorac Surg. 1999;68(1):208-211.
cept and design: Oxman, Marty, Panizales, McKenna, Frendl, 13. Duque JL, Ramos G, Castrodeza J, et al; Grupo Cooperativo de Carcinoma Bron- Bueno, Sugarbaker, and Baden. Acquisition of data: Ox- coge´nico de la Sociedad Espan˜ola de Neumologı´a y Cirugı´a Tora´cica. Early com-plications in surgical treatment of lung cancer: a prospective, multicenter study.
man, Issa, Patel, Panizales, Johnson, Licona, Frendl, Ann Thorac Surg. 1997;63(4):944-950.
Mentzer, Jaklitsch, Swanson, Sugarbaker, and Baden. Analy- 14. Hazelrigg SR, Magee MJ, Cetindag IB. Video-assisted thoracic surgery for diag- sis and interpretation of data: Oxman, Issa, Marty, Paniza- nosis of the solitary lung nodule. Chest Surg Clin N Am. 1998;8(4):763-774, vii.
les, Licona, Frendl, Colson, and Baden. Drafting of the manu- 15. Inderbitzi RG, Grillet MP. Risk and hazards of video-thoracoscopic surgery: a script: Oxman, Issa, Panizales, Frendl, Sugarbaker, and collective review. Eur J Cardiothorac Surg. 1996;10(7):483-489.
16. Jancovici R, Lang-Lazdunski L, Pons F, et al. Complications of video-assisted tho- Baden. Critical revision of the manuscript for important in- racic surgery: a five-year experience. Ann Thorac Surg. 1996;61(2):533-537.
tellectual content: Oxman, Issa, Marty, Patel, Johnson, Li- 17. Kaiser LR. Video-assisted thoracic surgery. Current state of the art. Ann Surg.
cona, McKenna, Frendl, Mentzer, Jaklitsch, Bueno, Col- son, Swanson, Sugarbaker, and Baden. Statistical analysis: 18. Kaiser LR, Bavaria JE. Complications of thoracoscopy. Ann Thorac Surg. 1993; Issa, Marty, and Panizales. Obtained funding: Frendl and 19. Krasna MJ, Deshmukh S, McLaughlin JS. Complications of thoracoscopy. Ann Baden. Administrative, technical, and material support: Marty, Thorac Surg. 1996;61(4):1066-1069.
Patel, Panizales, Johnson, Licona, McKenna, Frendl, Bueno, 20. Nan DN, Ferna´ndez-Ayala M, Farin˜as-Alvarez C, et al. Nosocomial infection after Sugarbaker, and Baden. Study supervision: Oxman, Issa, lung surgery: incidence and risk factors. Chest. 2005;128(4):2647-2652.
21. Rovera F, Imperatori A, Militello P, et al. Infections in 346 consecutive video- assisted thoracoscopic procedures. Surg Infect (Larchmt). 2003;4(1):45-51.
Conflicts of Interest Disclosures: None reported.
22. Ste´phan F, Boucheseiche S, Hollande J, et al. Pulmonary complications follow- Funding/Support: Funding for the study was provided by
ing lung resection: a comprehensive analysis of incidence and possible risk factors.
the Division of Infectious Diseases and the Surgical In- Chest. 2000;118(5):1263-1270.
tensive Care Unit Translational Research Center, Brigham 23. Tarkka M, Pokela R, Lepoja¨rvi M, Nissinen J, Ka¨rko¨la¨ P. Infection prophylaxis in pulmonary surgery: a randomized prospective study. Ann Thorac Surg. 1987; and Women’s Hospital, Boston, Massachusetts.
Role of the Sponsors: The Division of Infectious Dis-
24. Turna A, Kutlu CA, Ozalp T, Karamustafaoglu A, Mulazimog˘lu L, Bedirhan MA.
eases and the Surgical Intensive Care Unit Translational Antibiotic prophylaxis in elective thoracic surgery: cefuroxime versus cefepime.
Research Center did not participate in the design and con- Thorac Cardiovasc Surg. 2003;51(2):84-88.
duct of the study; in the collection, analysis, and inter- 25. Yim AP, Liu HP. Complications and failures of video-assisted thoracic surgery: experience from two centers in Asia. Ann Thorac Surg. 1996;61(2):538-541.
pretation of the data; or in the preparation, review, or 26. Engelman R, Shahian D, Shemin R, et al; Workforce on Evidence-Based Medi- cine, Society of Thoracic Surgeons. The Society of Thoracic Surgeons practice Additional Contributions: We thank Belisa B. Guzman-
guideline series: antibiotic prophylaxis in cardiac surgery, part II: antibiotic choice.
Suarez, MD, and Kimberly Cheng, BA, for their assis- Ann Thorac Surg. 2007;83(4):1569-1576.
27. Downey RJ. Complications after video-assisted thoracic surgery. Chest Surg Clin tance in the data collection and analysis and the Brigham and Women’s Hospital Investigational Drug Service team.
28. Craig SR, Hamzah M, Walker WS. Video-assisted thoracoscopic pneumonec- tomy for bronchial carcinoid tumor in a 14-year-old girl. J Pediatr Surg. 1996;31(12):1724-1726.
29. Walker WS, Craig SR. Video-assisted thoracoscopic pulmonary surgery—current status and potential evolution. Eur J Cardiothorac Surg. 1996;10(3):161-167.
1. Allen MS. Perioperative antibiotics: when, why? Thorac Surg Clin. 2005;15(2):229- 30. Walker WS, Pugh GC, Craig SR, Carnochan FM. Continued experience with tho- racoscopic major pulmonary resection. Int Surg. 1996;81(3):255-258.
2. LoCicero J III. Prophylactic antibiotic usage in cardiothoracic surgery. Chest. 1990; 31. Giudicelli R, Thomas P, Lonjon T, et al. Video-assisted minithoracotomy versus muscle-sparing thoracotomy for performing lobectomy. Ann Thorac Surg. 1994; 3. Bernard A, Pillet M, Goudet P, Viard H. Antibiotic prophylaxis in pulmonary sur- gery: a prospective randomized double-blind trial of flash cefuroxime versus forty- 32. Reimer K, Gleed C, Nicolle LE. The impact of postdischarge infection on surgical eight-hour cefuroxime. J Thorac Cardiovasc Surg. 1994;107(3):896-900.
wound infection rates. Infect Control. 1987;8(6):237-240.
4. Olak J, Jeyasingham K, Forrester-Wood C, Hutter J, al-Zeerah M, Brown E.
33. Manian FA, Meyer L. Comprehensive surveillance of surgical wound infections Randomized trial of one-dose versus six-dose cefazolin prophylaxis in elective in outpatient and inpatient surgery. Infect Control Hosp Epidemiol. 1990;11 general thoracic surgery. Ann Thorac Surg. 1991;51(6):956-958.
2013 American Medical Association. All rights reserved.
Downloaded From: http://archsurg.jamanetwork.com/ by a Harvard University User on 01/16/2013

Source: http://www.sleep_medicine_fellowship.jefferson.edu/content/dam/tju/jmc/files/medicine/Pulmonary_Critical_Care/news/jamasurg.2013.1372.pdf